A new artificial intelligence (AI) tool developed by a team at the University of Toronto may be able to significantly reduce the time needed to develop radiation therapy treatment plans for people with cancer.
The research published in the journal Medical Physics used AI to mine historical radiation therapy data and designed algorithms to develop recommended treatment strategies. To check the AI-produced relevant treatment plans, the researchers looked at 217 patients with head and neck cancer who had their radiation therapy schedules developed via conventional methods. The plans were comparable.
"There have been other AI optimization engines that have been developed, but the idea behind ours is that it more closely mimics the current clinical best practice," says Aaron Babier, the lead author of the research from the University of Toronto Engineering Department.
At the moment, developing radiation therapy plans for each individual patient's tumor can take days, valuable time for patients as the cancer often continues to grow and evolve, but also for physicians spending time designing these complex treatment strategies.
Head and neck cancers are notoriously difficult to design treatment plans for as tumors can be remarkably different from patient to patient. The researchers hope that as the tool worked so well on this tricky, complex cancer type, it should be able to handle more common tumor types that don't exhibit quite as much variation, such as prostate cancer.
Babier is keen to stress in this case that AI is not supposed to be a replacement for healthcare professionals, but may save them time by doing some important groundwork. Once the software has created a treatment plan, it would still be reviewed by a radiation physicist and further modified, taking at least a few more hours.
AI is touted to play a big role in the future of cancer diagnosis, monitoring and therapy, but concerns have been raised by some healthcare professionals about the ethics of using machine learning tools to make clinical decisions. One such concern published in an article earlier this year in the New England Journal of Medicine by researchers and MDs at Stanford stated:
'Physicians must adequately understand how algorithms are created, critically assess the source of the data used to create the statistical models designed to predict outcomes, understand how the models function and guard against becoming overly dependent on them.'
Herein lies a fairly common concern with new technological developments in medicine nowadays—the need for MDs to attain specialist knowledge about the new diagnostic methods they use so that they can fully understand how much to rely on them to influence their decisions about patients. A similar discussion is ongoing for perpetually-controversial liquid biopsies for cancer.
Despite these concerns, investment in AI from those in the healthcare industry is commonplace, with huge companies including Microsoft and IBM using it for various applications currently. Many firms seem to see AI as a possible solution to try to streamline the lengthy and obscenely costly drug-discovery process. Toronto-based biotech company BenchSci has as of today counted 28 pharmaceutical companies and 97 startups currently using AI for their drug-discovery processes.
In the case of using AI to aid radiation therapy treatment design, Babier says that his particular tool is more an extension of what is currently available to healthcare staff, rather than a revolution.
"It's essentially a pretty simple plugin to help with what is currently there in a clinical setting, but with more intelligent parameters than currently available," said Babier.
The University of Toronto team are not the only ones working on optimizing radiation therapy with AI. Other interested parties include Google's DeepMind Health, which is currently running a study with University College London Hospitals.
